Cleaning Knowledge Graphs - Talk by Davide Mottin (Aarhus University)
Автор: Immanuel Trummer
Загружено: 2022-03-29
Просмотров: 230
Описание: Is it possible to assess whether the information in a database is true? This task is an important tile in the current misinformation era. Towards this goal, the talk introduces the task of Knowledge validation taking knowledge graph, one of the data models that are used for fact-checking. Knowledge graphs are graphs of entities (e.g., US, Joe Biden) connected through relationships (e.g., isPresidentOf). A typical approach to ensure data quality in the presence of continuous changes is to apply logic rules. These rules are automatically mined from the data using frequency-based approaches. As a result, these approaches depend on the data quality of the KG and are susceptible to errors and incompleteness. This talk shows a different approach we recently devised in which the domain expert is a first-class citizen in the validation process. To ensure a smooth and easy validation we devised a method that selects triples to validate in the quest for cleaner data.
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
![Почему реактивный двигатель не плавится? [Veritasium]](https://imager.clipsaver.ru/F6UB0V2Fct4/max.jpg)
